iolanta 1.2.5__tar.gz → 1.2.6__tar.gz

{iolanta-1.2.5 → iolanta-1.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: iolanta
-Version: 1.2.5
+Version: 1.2.6
 Summary: Semantic Web browser
 License: MIT
 Author: Anatoly Scherbakov
@@ -15,12 +15,14 @@ Provides-Extra: all
 Requires-Dist: boltons (>=24.0.0)
 Requires-Dist: classes (>=0.4.0)
 Requires-Dist: deepmerge (>=0.1.1)
+Requires-Dist: diskcache (>=5.6.3)
 Requires-Dist: documented (>=0.1.1)
 Requires-Dist: dominate (>=2.6.0)
 Requires-Dist: funcy (>=2.0)
 Requires-Dist: iolanta-tables (>=0.1.7,<0.2.0) ; extra == "all"
 Requires-Dist: loguru (>=0.7.3)
 Requires-Dist: more-itertools (>=9.0.0)
+Requires-Dist: nanopub (>=2.0.1)
 Requires-Dist: owlrl (>=6.0.2)
 Requires-Dist: oxrdflib (>=0.4.0)
 Requires-Dist: python-frontmatter (>=0.5.0)

{iolanta-1.2.5 → iolanta-1.2.6}/iolanta/cyberspace/processor.py

@@ -8,14 +8,18 @@ from threading import Lock
 from types import MappingProxyType
 from typing import Any, Iterable, Mapping
 
+import diskcache
 import funcy
 import loguru
+import platformdirs
 import reasonable
 import yaml_ld
+from nanopub import NanopubClient
 from rdflib import ConjunctiveGraph, Namespace, URIRef, Variable
 from rdflib.plugins.sparql.algebra import translateQuery
 from rdflib.plugins.sparql.evaluate import evalQuery
 from rdflib.plugins.sparql.parser import parseQuery
+from rdflib.plugins.sparql.parserutils import CompValue
 from rdflib.plugins.sparql.sparql import Query
 from rdflib.query import Processor
 from rdflib.term import BNode, Literal, Node
@@ -53,7 +57,7 @@ REDIRECTS = MappingProxyType({
     #   - either find a way to resolve these URLs automatically,
     #   - or create a repository of those redirects online.
     'https://purl.org/vocab/vann/': URIRef(
-        'https://vocab.org/vann/vann-vocab-20100607.rdf'
+        'https://vocab.org/vann/vann-vocab-20100607.rdf',
     ),
     URIRef(DC): URIRef(DCTERMS),
     URIRef(RDF): URIRef(RDF),
@@ -69,6 +73,34 @@ REDIRECTS = MappingProxyType({
 })
 
 
+@diskcache.Cache(
+    directory=str(
+        platformdirs.user_cache_path(
+            appname='iolanta',
+        ) / 'find_retractions_for',
+    ),
+).memoize(expire=datetime.timedelta(days=8).total_seconds())
+def find_retractions_for(nanopublication: URIRef) -> set[URIRef]:
+    """Find nanopublications that retract the given one."""
+    # See https://github.com/fair-workflows/nanopub/issues/168 for
+    # context of this dirty hack.
+    use_server = 'http://grlc.nanopubs.lod.labs.vu.nl/api/local/local/'
+
+    client = NanopubClient(
+        use_server=use_server,
+    )
+    client.grlc_urls = [use_server]
+
+    retractions = client.find_retractions_of(
+        str(nanopublication).replace(
+            'https://',
+            'http://',
+        ),
+    )
+
+    return {URIRef(retraction) for retraction in retractions}
+
+
 def _extract_from_mapping(  # noqa: WPS213
     algebra: Mapping[str, Any],
 ) -> Iterable[URIRef | Variable]:
@@ -188,15 +220,12 @@ def resolve_variables(
 
 
 def extract_mentioned_urls_from_query(
-    query: str,
+    query: Query,
     bindings: dict[str, Node],
     base: str | None,
     namespaces: dict[str, Namespace],
 ) -> tuple[Query, set[URIRef]]:
     """Extract URLs that a SPARQL query somehow mentions."""
-    parse_tree = parseQuery(query)
-    query = translateQuery(parse_tree, base, namespaces)
-
     return query, set(
         resolve_variables(
             extract_mentioned_urls(query.algebra),
@@ -218,6 +247,35 @@ class Skipped:
 LoadResult = Loaded | Skipped
 
 
+def _extract_nanopublication_uris(
+    algebra: CompValue,
+) -> Iterable[URIRef]:
+    """Extract nanopublications to get retracting information for."""
+    match algebra.name:
+        case 'SelectQuery' | 'Project' | 'Distinct' | 'Graph':
+            yield from _extract_nanopublication_uris(algebra['p'])
+
+        case 'BGP':
+            for retractor, retracts, retractee in algebra['triples']:
+                if retracts == URIRef(
+                    'https://purl.org/nanopub/x/retracts',
+                ) and isinstance(retractor, Variable):
+                    yield retractee
+
+        case 'LeftJoin' | 'Join':
+            yield from _extract_nanopublication_uris(algebra['p1'])
+            yield from _extract_nanopublication_uris(algebra['p2'])
+
+        case 'Filter' | 'OrderBy':
+            return
+
+        case unknown_name:
+            raise ValueError(
+                f'Unknown algebra name: {unknown_name}, '
+                f'content: {algebra}',
+            )
+
+
 @dataclasses.dataclass(frozen=True)
 class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
     """
@@ -325,8 +383,18 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
             query = strOrQuery
 
         else:
+            parse_tree = parseQuery(strOrQuery)
+            query = translateQuery(parse_tree, base, initNs)
+
+            self.load_retracting_nanopublications_by_query(
+                query=query,
+                bindings=initBindings,
+                base=base,
+                namespaces=initNs,
+            )
+
             query, urls = extract_mentioned_urls_from_query(
-                query=strOrQuery,
+                query=query,
                 bindings=initBindings,
                 base=base,
                 namespaces=initNs,
@@ -441,7 +509,7 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
             return Loaded()
 
         except Exception as err:
-            self.logger.info('%s | Failed: %s', source, err)
+            self.logger.info(f'{source} | Failed: {err}')
 
             self.graph.add((
                 source_uri,
@@ -573,3 +641,32 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
             )
 
         return term
+
+    def load_retracting_nanopublications_by_query(    # noqa: WPS231
+        self,
+        query: Query,
+        bindings: dict[str, Node],
+        base: str | None,
+        namespaces: dict[str, Namespace],
+    ):
+        """
+        If the query requires information about retracting nanopubs, load them.
+
+        FIXME: This function presently does nothing because `nanopub` library
+          has problems: https://github.com/fair-workflows/nanopub/issues/168
+
+        TODO: Generalize this mechanism to allow for plugins which analyze
+          SPARQL queries and load information into the graph based on their
+          content.
+        """
+        nanopublications = list(
+            resolve_variables(
+                terms=_extract_nanopublication_uris(query.algebra),
+                bindings=bindings,
+            ),
+        )
+
+        for nanopublication in nanopublications:
+            retractions = find_retractions_for(nanopublication)
+            for retraction in retractions:
+                self.load(retraction)

{iolanta-1.2.5 → iolanta-1.2.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "iolanta"
-version = "1.2.5"
+version = "1.2.6"
 description = "Semantic Web browser"
 authors = ["Anatoly Scherbakov <altaisoft@gmail.com>"]
 license = "MIT"
@@ -33,6 +33,8 @@ yaml-ld = ">=1.1.3"
 reasonable = ">=0.2.6"
 oxrdflib = ">=0.4.0"
 loguru = ">=0.7.3"
+nanopub = ">=2.0.1"
+diskcache = ">=5.6.3"
 
 [tool.poetry.scripts]
 iolanta = "iolanta.cli:app"